Linguaphile wrote:I'd noticed this before and wondered: for some meanings there is just one word derived from one of the four languages and for others there are two different words or even four, like for the translation of "cat" above. I'm curious whether there is any specific reason for this. Of course we have synonyms that are derived from different languages in English too: to use the "forever" and "cat" examples again, there's forever and eternal and perpetual, cat and feline and kitty and so on. So it's not a surprising feature, but it's kind of fun to see it in your conlang.
How did you decide which words would have more than one form? Or is it fairly random, since that's probably how it would appear in a natural language of this type? It seems like it would be a difficult feature to create in a way that seems natural, but the way you've done it actually does - most words with just one form, some words with two forms, a few with three, fewer with four - avoiding the temptation to just choose one word for each meaning (which is what I actually did, when I created a vocabulary for a mixed language like this years ago; I didn't think of doing it like you did) or the temptation to create four synonyms for each meaning. In a natural language it would more likely be like what you've done, some words with one form, some with two, three, or four. So I'm curious if there was a method to how you chose which words would have multiple forms, or if it was just random. (I can think of a way to not leave it to chance, but I don't know if it's what you've done.)
The explanation is simple: the four languages (Hanuvaa, Tuuruha, "third language", and "fourth language") that form the Tunuvaa language were created by a program that creates vocabulary. You can create more than one vocabulary with the same set of phonemes. This is what I did. But the word list that the program generates is not fixed: sometimes a term appears in the word list, other times that same term does not appear. I decided only which phonemes the languages should contain, then let the program create the words. I discarded the list of words with phonotactics that I did not like. In the end I got four languages that all shared the same phonotactic structure, which was pleasant to my ears. In the case of the language I later named "Hanuvaa," I added one more phoneme than in the other three languages (the /g/ phoneme). Sometimes, by chance, in two lists, or in three, or even in all four list of words there was the same word. Initially I wanted to work with one language, the one that sounded best to my ear (it was the Hanuvaa language), but then I decided that I could not throw away the second language, which I called Tuuruha. Later, I decided that I had to use all four languages, because their phonetic structure was so fascinating to me and they had the same phonotactics. So I decided to put all four vocabularies generated by the program together in alphabetical order, thus forming one language. This is the reason for the presence, in the resulting final list, of up to four words for the same concept.
Then I invented the story of the Tuuruha-speaking people, which were fluent also in other two languages, and later adopted also the Hanuvaa language. I still have to understand if this historical reconstruction is plausible. Maybe I will discard some synonims, or change the meaning of some of them, in order to have a list of words with different connontations.